Object detection may be utilized to support a wide variety of functionality. One example of this functionality is face detection. Face detection may be utilized in a variety of ways, such as to determine whether an image includes a face, an identity of a face that is included in the image (e.g., facial recognition), a location of the face in the image, the scale and size of a face, the orientation of a face, and so on.
Conventional techniques that were utilized to perform face detection, however, were often resource intensive. Therefore, these techniques were off ill-suited for use by mobile communications devices that may have limited processing or memory resources such as mobile phones, tablets and so forth. For example, these techniques, when utilized by a mobile communications device, could consume a majority of the available computational resources of the device, may not support performance in real time, drain battery quickly and so on and therefore limit the usefulness of the techniques with the device.
Consequently, use of these techniques was often limited to devices having a significant amount of resources. Even then, the proliferation of techniques that are utilized to conserve power and even computational resources of resource-rich devices could also limit the usefulness of conventional face detection techniques from being performed adequately by these devices.